Distributed data processing system, device and method

Abstract

The invention proposes a distributed data processing system, equipment and method. Wherein, the method includes: allocating each of at least one data processing task to one of at least one computing node server, wherein allocating the at least one computing node server based on the performance level of each of the at least one computing node server Data processing tasks; each of the at least one computing node server executes the assigned data processing tasks, and transmits the data processing results to the result management server; the result management server uniformly manages the data processing results. The distributed data processing system, equipment and method disclosed in the invention have high data processing performance, low cost and can uniformly manage data processing results.

Description

technical field [0001] The present invention relates to data processing systems, devices and methods, and more particularly, to distributed data processing systems, devices and methods. Background technique [0002] At present, with the widening of computer applications and the increasing variety of business types in different fields, it becomes more and more important to perform data processing in a distributed manner through the cooperative work of several computing devices. [0003] Existing distributed data processing systems and methods usually include a task management server and at least one computing node server. The basic working principle is as follows: the task management server allocates computing tasks to each computing node server; each computing node server executes the received data. The computing task; after the single-node server completes the computing task, the computing results and parameters are stored in the node. [0004] However, the existing distri...

AI Technical Summary

Problems solved by technology

[0004] However, the existing distributed data processing systems and methods have the following problems: Since each computing node server has both management and computing functions, that is, system management and users model and compute in a single node, it affects computing performance And it is not conducive to the rapid application of calculation results. In addition, when it is necessary to add computing node servers, it is necessary to reconfigure the newly added nodes (such as reconfiguring the database, operating system, parameter environment, etc.), so it is cumbersome to use. Computing node servers are deployed with the same data backup, so that resource utilization is low. In addition, the existing distributed data processing systems and methods further have the following problems: when a single node computing times out or fails, the same computing needs to be deployed on a new node. , resulting in a decrease in system computing efficiency. At the same time, since the calculation results of each node are stored in their own nodes, it is difficult to manage them in a unified manner. In addition, due to the single-node calculation method, the hardware configuration requirements for each node are relatively high. , resulting in a higher overall system cost

Images